slcom/slangc/parser/SourceEncoding.sauce

package slangc.parser;

import slang.vm.SystemCall;

public class SourceEncoding {

	public SourceEncoding() {
		// TODO Auto-generated constructor stub
	}
	
	public String getEncodingName() {
		return "Generic Latin/Unicode+Hieroglyphs";
	}
	
	public boolean isNewline(int ch) {
		return ch == '\n';
	}

	public boolean isSpace(int ch) {
		return ch == ' ' || ch == '\t' || ch == '\f' || ch == '\r' || isNewline(ch);
	}
	
	public boolean isDigit(int ch) {
		return isDigit(ch, 10);
	}
	
	public boolean isDigitOrUnderscore(int ch) {
		return isDigit(ch) || ch == '_';
	}
	
	public boolean isDigit(int ch, int base) {
		if (base != 2 && base != 10 && base != 16) {
			throw new Error("Bad base (expected 10 or 16): " + base);
		}
		
		if (ch == '0' || ch == '1' || ch == 0x660/*'٠'*/ || ch == 0x661/*'١'*/) {
			return true;
		} else if ((ch >= '2' && ch <= '9') || (ch >= 0x662/*'٢'*/ && ch <= 0x669 /*'٩'*/)) {
			return base >= 10;
		} else if ((ch >= 'a' && ch <= 'f') || (ch >= 'A' && ch <= 'F')) {
			return base >= 16;
		} else if (ch == 'ⲁ' || ch == 'ⲃ' || ch == 'ⲅ' || ch == 'ⲇ' || ch == 'ⲋ' || ch == 'ⲉ') {
			return base >= 16;
		} else {
			return false;
		}
	}
	
	public boolean isDigitOrUnderscore(int ch, int base) {
		return isDigit(ch, base) || ch == '_';
	}
	
	public boolean isDecimalPoint(int ch) {
		return ch == '.';
	}
	
	public boolean isExponentSymbol(int ch) {
		return ch == 'e' || ch == 'E' || ch == 'p' || ch == 'P'; // p/P forms used in hex floats
	}
	
	public boolean isSign(int ch) {
		return ch == '+' || ch == '-';
	}
	
	public boolean isIntegerSuffix(int ch) {
		return ch == 'L' || ch == 'l' || ch == 'U' || ch == 'u';
	}
	
	public boolean isFloatSuffix(int ch) {
		return ch == 'F' || ch == 'f' || ch == 'D' || ch == 'd';
	}
	
	public boolean isAlphaNubian(int ch) {
		switch (ch) {
		case 'ⲁ':
		case 'ⲃ':
		case 'ⲅ':
		case 'ⲇ':
		case 'ⲉ':
		case 'ⲋ':
		case 'ⲍ':
		case 'ⲏ':
		case 'ⲑ':
		case 'ⲓ':
		case 'ⲕ':
		case 'ⲗ':
		case 'ⲙ':
		case 'ⲛ':
		case 'ⲝ':
		case 'ⲟ':
		case 'ⲡ':
		case 'ⲣ':
		case 'ⲥ':
		case 'ⲧ':
		case 'ⲩ':
		case 'ⲫ':
		case 'ⲭ':
		case 'ⲯ':
		case 'ⲱ':
		case 'ϣ':
		case 'ⳡ'://ⳡⲁⲓ??
		//case 'ⲁ'://ϩⲁⲓ???
		case 'ϩ':
		case 'ⳝ':
		//case 'ⲧ̄'://ⲧⲓ??
		//case ''://??
		case 'ⳟ':
		//case 'ⲁ': //ⲛⲓ?
		//ⲱⲁⲱ?
			return true;
		default:
			return false;
		}
	}
	
	public boolean isAlphabetical(int ch) {
		return (SystemCall.characterFlags(ch) & 1) == 1;
		throw new Error("TODO: Alphabetical handling"); //return Character.isAlphabetic(ch);//(ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z');
	}
	
	public boolean isValidNameHead(int ch) {
		return isAlphabetical(ch) || ch == '_' || ch == '$'
				/* Cleopatra said we're allowed to use hieroglyphs. Ase. */
				|| (ch >= 0x13000 && ch <= 0x13430) || ch == 0x2625 // TODO: Fix hex support (breaks if ends with F/D due to confusion with float/double tags?)
				/* Nubian was my call. I wanted it in the first version.*/
				|| isAlphaNubian(ch);
	}
	
	public boolean isValidNameTail(int ch) {
		return isValidNameHead(ch) || isDigit(ch);
	}
	
	public String getString(int ch) {
		return String.construct(new int[]{ch});
	}
}